setwd("~/Documents/GitHub/Resultados/docs/PrimeroDiffExpAllResults/Clasificando/Abundances")
load('CheackPointOne.RData')
head(pEhExvsCDC5,10);
## GenId CDC5_1 CDC5_2 CDC5_3 pEhEx_1 pEhEx_2 pEhEx_3
## 1 EHI_000130A 72.594848 40.99764 280.54541 169.92967 411.34855 466.86626
## 2 EHI_000140A 111.027415 145.24877 643.77044 342.73950 324.16054 37.85402
## 3 EHI_000240A 876.831894 752.01378 1122.88829 1140.54489 1837.65493 2071.45616
## 4 EHI_000250A 492.506226 510.71341 232.49229 315.37794 310.74700 642.81736
## 5 EHI_000260A 12.810856 107.76521 2.11999 105.84602 63.71431 49.07003
## 6 EHI_000280A 58.360564 42.16900 171.71923 53.28303 79.36344 60.28603
## 7 EHI_000290A 17.081141 26.94130 14.13327 18.00102 13.41354 79.21304
## 8 EHI_000300A 49.819994 70.28166 48.05312 129.60737 109.54391 30.84402
## 9 EHI_000410A 14.234284 19.91314 52.29310 27.36156 23.47369 63.79104
## 10 EHI_000430A 9.963999 25.76994 2.11999 25.20143 13.41354 10.51501
nbreaks <- 10
data1 <- pEhExvsCDC5; head(data1)
## GenId CDC5_1 CDC5_2 CDC5_3 pEhEx_1 pEhEx_2 pEhEx_3
## 1 EHI_000130A 72.59485 40.99764 280.54541 169.92967 411.34855 466.86626
## 2 EHI_000140A 111.02742 145.24877 643.77044 342.73950 324.16054 37.85402
## 3 EHI_000240A 876.83189 752.01378 1122.88829 1140.54489 1837.65493 2071.45616
## 4 EHI_000250A 492.50623 510.71341 232.49229 315.37794 310.74700 642.81736
## 5 EHI_000260A 12.81086 107.76521 2.11999 105.84602 63.71431 49.07003
## 6 EHI_000280A 58.36056 42.16900 171.71923 53.28303 79.36344 60.28603
sample1 <- data1$pEhEx_1; sample2 <- data1$pEhEx_2; sample3 <- data1$pEhEx_3;
samplevs1 <- data1$CDC5_1; samplevs2 <- data1$CDC5_2; samplevs3 <- data1$CDC5_3;
log2sample1 <- log2(sample1+1); log2sample2 <- log2(sample2+1)
log2sample3 <- log2(sample3+1); log2samplevsCDC51 <- log2(samplevs1+1)
log2samplevsCDC52 <- log2(samplevs2+1); log2samplevsCDC53 <- log2(samplevs3+1)
data1 <- cbind(data1, log2sample1,log2sample2,log2sample3,
log2samplevsCDC51,log2samplevsCDC52,log2samplevsCDC53)
head(data1)
## GenId CDC5_1 CDC5_2 CDC5_3 pEhEx_1 pEhEx_2 pEhEx_3
## 1 EHI_000130A 72.59485 40.99764 280.54541 169.92967 411.34855 466.86626
## 2 EHI_000140A 111.02742 145.24877 643.77044 342.73950 324.16054 37.85402
## 3 EHI_000240A 876.83189 752.01378 1122.88829 1140.54489 1837.65493 2071.45616
## 4 EHI_000250A 492.50623 510.71341 232.49229 315.37794 310.74700 642.81736
## 5 EHI_000260A 12.81086 107.76521 2.11999 105.84602 63.71431 49.07003
## 6 EHI_000280A 58.36056 42.16900 171.71923 53.28303 79.36344 60.28603
## log2sample1 log2sample2 log2sample3 log2samplevsCDC51 log2samplevsCDC52
## 1 7.417259 8.687721 8.869952 6.201533 5.392236
## 2 8.425172 8.345008 5.279992 6.807708 7.192281
## 3 10.156772 10.844435 11.017126 9.777801 9.556532
## 4 8.305505 8.284232 9.330508 8.946924 8.999192
## 5 6.739389 6.016013 5.645875 3.787731 6.765073
## 6 5.762429 6.328467 5.937486 5.891433 5.431924
## log2samplevsCDC53
## 1 8.137224
## 2 9.332642
## 3 10.134283
## 4 7.867231
## 5 1.641542
## 6 7.432285
save.image('CheckPointTwo.RData')
setwd("~/Documents/GitHub/Resultados/docs/PrimeroDiffExpAllResults/Clasificando/Abundances")
#load('CheckPointTwo.RData')
library(ggplot2);library(dplyr);library("fitdistrplus");
##
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
##
## filter, lag
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
## Loading required package: MASS
##
## Attaching package: 'MASS'
## The following object is masked from 'package:dplyr':
##
## select
## Loading required package: survival
library("MASS");library("survival")
head(data1)
## GenId CDC5_1 CDC5_2 CDC5_3 pEhEx_1 pEhEx_2 pEhEx_3
## 1 EHI_000130A 72.59485 40.99764 280.54541 169.92967 411.34855 466.86626
## 2 EHI_000140A 111.02742 145.24877 643.77044 342.73950 324.16054 37.85402
## 3 EHI_000240A 876.83189 752.01378 1122.88829 1140.54489 1837.65493 2071.45616
## 4 EHI_000250A 492.50623 510.71341 232.49229 315.37794 310.74700 642.81736
## 5 EHI_000260A 12.81086 107.76521 2.11999 105.84602 63.71431 49.07003
## 6 EHI_000280A 58.36056 42.16900 171.71923 53.28303 79.36344 60.28603
## log2sample1 log2sample2 log2sample3 log2samplevsCDC51 log2samplevsCDC52
## 1 7.417259 8.687721 8.869952 6.201533 5.392236
## 2 8.425172 8.345008 5.279992 6.807708 7.192281
## 3 10.156772 10.844435 11.017126 9.777801 9.556532
## 4 8.305505 8.284232 9.330508 8.946924 8.999192
## 5 6.739389 6.016013 5.645875 3.787731 6.765073
## 6 5.762429 6.328467 5.937486 5.891433 5.431924
## log2samplevsCDC53
## 1 8.137224
## 2 9.332642
## 3 10.134283
## 4 7.867231
## 5 1.641542
## 6 7.432285
log2sample1 <- data1$log2sample1; head(mean(log2sample1)); head(sd(log2sample1))
## [1] 6.30237
## [1] 2.868113
head(log2sample1,5)
## [1] 7.417259 8.425172 10.156772 8.305505 6.739389
summary(data1)
## GenId CDC5_1 CDC5_2 CDC5_3
## Length:4772 Min. : 0.00 Min. : 0.00 Min. : 0.0
## Class :character 1st Qu.: 17.08 1st Qu.: 17.57 1st Qu.: 16.3
## Mode :character Median : 45.55 Median : 49.20 Median : 44.5
## Mean : 1749.28 Mean : 1748.01 Mean : 1980.0
## 3rd Qu.: 196.79 3rd Qu.: 208.50 3rd Qu.: 177.4
## Max. :270953.87 Max. :270338.41 Max. :481876.0
## pEhEx_1 pEhEx_2 pEhEx_3 log2sample1
## Min. : 0.0 Min. : 0.00 Min. : 0.0 Min. : 0.000
## 1st Qu.: 18.0 1st Qu.: 15.65 1st Qu.: 15.4 1st Qu.: 4.248
## Median : 50.4 Median : 49.18 Median : 54.0 Median : 5.684
## Mean : 1395.0 Mean : 1717.64 Mean : 1909.2 Mean : 6.302
## 3rd Qu.: 208.1 3rd Qu.: 223.84 3rd Qu.: 242.0 3rd Qu.: 7.708
## Max. :207266.7 Max. :265749.05 Max. :707261.7 Max. :17.661
## log2sample2 log2sample3 log2samplevsCDC51 log2samplevsCDC52
## Min. : 0.000 Min. : 0.000 Min. : 0.000 Min. : 0.000
## 1st Qu.: 4.057 1st Qu.: 4.038 1st Qu.: 4.176 1st Qu.: 4.215
## Median : 5.649 Median : 5.781 Median : 5.541 Median : 5.650
## Mean : 6.237 Mean : 6.186 Mean : 6.244 Mean : 6.270
## 3rd Qu.: 7.813 3rd Qu.: 7.925 3rd Qu.: 7.628 3rd Qu.: 7.711
## Max. :18.020 Max. :19.432 Max. :18.048 Max. :18.044
## log2samplevsCDC53
## Min. : 0.000
## 1st Qu.: 4.109
## Median : 5.508
## Mean : 6.114
## 3rd Qu.: 7.479
## Max. :18.878
ndata1 <- length(log2sample1)
hist(log2sample1, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2 sample1')
meanlog2sample1 <- mean(log2sample1); head(meanlog2sample1)
## [1] 6.30237
StdDevlog2sample1 <- sd(log2sample1); head(StdDevlog2sample1)
## [1] 2.868113
Normlog2sample1 <- (log2sample1-meanlog2sample1)/StdDevlog2sample1; head(Normlog2sample1)
## [1] 0.3887185 0.7401387 1.3438805 0.6984156 0.1523717 -0.1882564
tst<- Normlog2sample1
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2sample1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -9.053474e-17 0.01447452
## sd 9.998952e-01 0.01023499
## Loglikelihood: -6770.675 AIC: 13545.35 BIC: 13558.29
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
Cálculo de cuantiles
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8713278 0.8126272
## 70 -0.9260903 0.9617987
## 75 -0.9875514 1.1301158
## 80 -1.0213487 1.3734820
## 85 -1.0966280 1.6988689
## 90 -1.1851905 2.1819106
## 95 -1.3565981 2.6079805
## 99 -1.9245847 3.1666752
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 BaseMean - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2sample2 <- data1$log2sample2; head(mean(log2sample2)); head(sd(log2sample2))
## [1] 6.237216
## [1] 3.103083
head(log2sample2,5)
## [1] 8.687721 8.345008 10.844435 8.284232 6.016013
ndata1 <- length(log2sample2)
hist(log2sample2, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2 sample2')
Log-normalizacion
meanlog2sample2 <- mean(log2sample2); head(meanlog2sample2)
## [1] 6.237216
StdDevlog2sample2 <- sd(log2sample2); head(StdDevlog2sample2)
## [1] 3.103083
Normlog2sample2 <- (log2sample2-meanlog2sample2)/StdDevlog2sample2; head(Normlog2sample2)
## [1] 0.78969998 0.67925752 1.48472313 0.65967167 -0.07128491 0.02940674
tst<- Normlog2sample2
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2sample1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajuste de modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -1.260395e-16 0.01447452
## sd 9.998952e-01 0.01023499
## Loglikelihood: -6770.675 AIC: 13545.35 BIC: 13558.29
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8926336 0.8466541
## 70 -0.9421687 0.9875692
## 75 -0.9976178 1.1428700
## 80 -1.0605870 1.3752461
## 85 -1.1334411 1.6895575
## 90 -1.2198720 2.1202686
## 95 -1.4640882 2.5136403
## 99 -2.0100061 3.0865078
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 2)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 BaseMean - ADJUSTED (sample 2)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 2)', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2sample3 <- data1$log2sample3; head(mean(log2sample3)); head(sd(log2sample3))
## [1] 6.186357
## [1] 3.171257
head(log2sample3,5)
## [1] 8.869952 5.279992 11.017126 9.330508 5.645875
ndata1 <- length(log2sample3)
hist(log2sample3, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2 sample2')
meanlog2sample3 <- mean(log2sample3); head(meanlog2sample3)
## [1] 6.186357
StdDevlog2sample3 <- sd(log2sample3); head(StdDevlog2sample3)
## [1] 3.171257
Normlog2sample3 <- (log2sample3-meanlog2sample3)/StdDevlog2sample3; head(Normlog2sample3)
## [1] 0.84622446 -0.28580629 1.52329778 0.99145245 -0.17043143 -0.07847701
tst<- Normlog2sample3
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2sample1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando Modelos
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean 2.945427e-17 0.01447452
## sd 9.998952e-01 0.01023499
## Loglikelihood: -6770.675 AIC: 13545.35 BIC: 13558.29
## Correlation matrix:
## mean sd
## mean 1.000000e+00 1.684231e-11
## sd 1.684231e-11 1.000000e+00
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8676459 0.8379706
## 70 -0.9660265 0.9701830
## 75 -1.0458574 1.1161493
## 80 -1.1427428 1.3265284
## 85 -1.2660057 1.6213606
## 90 -1.4356115 2.0178765
## 95 -1.7090932 2.4795401
## 99 -1.9507587 3.0829063
Creación de histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 3)', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 3)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 BaseMean - ADJUSTED (sample 3)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 3)', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 3)', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2vsCDC51 <- data1$log2samplevsCDC51; head(mean(log2vsCDC51)); head(sd(log2vsCDC51))
## [1] 6.244372
## [1] 2.880412
head(log2vsCDC51,5)
## [1] 6.201533 6.807708 9.777801 8.946924 3.787731
ndata1 <- length(log2vsCDC51)
hist(log2vsCDC51, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2vsCDC51')
meanlog2vsCDC51 <- mean(log2vsCDC51); head(meanlog2vsCDC51)
## [1] 6.244372
StdDevlog2vsCDC51 <- sd(log2vsCDC51); head(StdDevlog2vsCDC51)
## [1] 2.880412
Normlog2vsCDC51 <- (log2vsCDC51-meanlog2vsCDC51)/StdDevlog2vsCDC51; head(Normlog2vsCDC51)
## [1] -0.01487273 0.19557464 1.22670936 0.93825188 -0.85287855 -0.12253093
tst<- Normlog2vsCDC51
Primer histograma
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2vsCDC51',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean 6.276949e-17 0.01447452
## sd 9.998952e-01 0.01023499
## Loglikelihood: -6770.675 AIC: 13545.35 BIC: 13558.29
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8528786 0.8216616
## 70 -0.9073590 0.9496778
## 75 -0.9073590 1.1007491
## 80 -0.9684970 1.3308952
## 85 -1.0381490 1.6867728
## 90 -1.1190762 2.1306178
## 95 -1.2156475 2.6928135
## 99 -2.1678747 3.3316192
Histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2vsCDC52 <- data1$log2samplevsCDC52; head(mean(log2vsCDC52)); head(sd(log2vsCDC52))
## [1] 6.269726
## [1] 2.952289
head(log2vsCDC52,5)
## [1] 5.392236 7.192281 9.556532 8.999192 6.765073
Primer Histograma
ndata1 <- length(log2vsCDC52)
hist(log2vsCDC52, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2vsCDC52')
meanlog2vsCDC52 <- mean(log2vsCDC52); head(meanlog2vsCDC52)
## [1] 6.269726
StdDevlog2vsCDC52 <- sd(log2vsCDC52); head(StdDevlog2vsCDC52)
## [1] 2.952289
Normlog2vsCDC52 <- (log2vsCDC52-meanlog2vsCDC52)/StdDevlog2vsCDC52; head(Normlog2vsCDC52)
## [1] -0.2972235 0.3124880 1.1133080 0.9245256 0.1677843 -0.2837806
tst<- Normlog2vsCDC52
** Segundo Histograma**
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2vsCDC52',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -3.587454e-18 0.01447452
## sd 9.998952e-01 0.01023499
## Loglikelihood: -6770.675 AIC: 13545.35 BIC: 13558.29
## Correlation matrix:
## mean sd
## mean 1.000000e+00 1.684231e-11
## sd 1.684231e-11 1.000000e+00
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
Cálculo de cuantiles
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8380866 0.7932340
## 70 -0.8811551 0.9359679
## 75 -0.9283894 1.1117853
## 80 -1.0392501 1.3412158
## 85 -1.1058050 1.6629743
## 90 -1.1828753 2.0953140
## 95 -1.3871587 2.6382746
## 99 -2.1236833 3.2437321
Histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2vsCDC53 <- data1$log2samplevsCDC53; head(mean(log2vsCDC53)); head(sd(log2vsCDC53))
## [1] 6.11433
## [1] 2.904448
head(log2vsCDC53,5)
## [1] 8.137224 9.332642 10.134283 7.867231 1.641542
ndata1 <- length(log2vsCDC53)
** Primer histograma**
hist(log2vsCDC53, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2vsCDC53')
meanlog2vsCDC53 <- mean(log2vsCDC53); head(meanlog2vsCDC53)
## [1] 6.11433
StdDevlog2vsCDC53 <- sd(log2vsCDC53); head(StdDevlog2vsCDC53)
## [1] 2.904448
Normlog2vsCDC53 <- (log2vsCDC53-meanlog2vsCDC53)/StdDevlog2vsCDC53; head(Normlog2vsCDC53)
## [1] 0.6964814 1.1080632 1.3840678 0.6035230 -1.5399785 0.4537713
tst<- Normlog2vsCDC53
Segundo histograma
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2vsCDC53',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -9.490135e-17 0.01447452
## sd 9.998952e-01 0.01023499
## Loglikelihood: -6770.675 AIC: 13545.35 BIC: 13558.29
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
Calculo de cuantiles
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8305939 0.8018512
## 70 -0.8877013 0.9412276
## 75 -0.9522378 1.1212912
## 80 -1.0264291 1.3220406
## 85 -1.1638210 1.5939827
## 90 -1.2824356 2.0684839
## 95 -1.4385679 2.6057307
## 99 -1.8396440 3.5572717
Histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
head(pEhExvsCmasM,10);
## # A tibble: 10 × 7
## GenId CDC5_1 CDC5_2 CDC5_3 pEhEx_1 pEhEx_2 pEhEx_3
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 EHI_000130A 45.3 108. 129. 180. 446. 516.
## 2 EHI_000140A 66.0 318. 257. 363. 352. 41.8
## 3 EHI_000240A 701. 1282. 877. 1209. 1993. 2290.
## 4 EHI_000250A 707. 430. 389. 334. 337. 711.
## 5 EHI_000260A 94.5 109. 35.2 112. 69.1 54.2
## 6 EHI_000280A 58.3 50.5 80.8 56.5 86.1 66.6
## 7 EHI_000290A 27.2 14.9 23.9 19.1 14.5 87.6
## 8 EHI_000300A 60.9 143. 111. 137. 119. 34.1
## 9 EHI_000410A 15.5 21.8 23.2 29.0 25.5 70.5
## 10 EHI_000430A 27.2 27.5 22.4 26.7 14.5 11.6
nbreaks <- 10
data2 <- pEhExvsCmasM; head(data2)
## # A tibble: 6 × 7
## GenId CDC5_1 CDC5_2 CDC5_3 pEhEx_1 pEhEx_2 pEhEx_3
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 EHI_000130A 45.3 108. 129. 180. 446. 516.
## 2 EHI_000140A 66.0 318. 257. 363. 352. 41.8
## 3 EHI_000240A 701. 1282. 877. 1209. 1993. 2290.
## 4 EHI_000250A 707. 430. 389. 334. 337. 711.
## 5 EHI_000260A 94.5 109. 35.2 112. 69.1 54.2
## 6 EHI_000280A 58.3 50.5 80.8 56.5 86.1 66.6
sample1 <- data2$pEhEx_1; sample2 <- data2$pEhEx_2; sample3 <- data2$pEhEx_3;
samplevs1 <- data2$CDC5_1; samplevs2 <- data2$CDC5_2; samplevs3 <- data2$CDC5_3;
log2sample1 <- log2(sample1+1); log2sample2 <- log2(sample2+1)
log2sample3 <- log2(sample3+1); log2samplevsCDC51 <- log2(samplevs1+1)
log2samplevsCDC52 <- log2(samplevs2+1); log2samplevsCDC53 <- log2(samplevs3+1)
data2 <- cbind(data2, log2sample1,log2sample2,log2sample3,
log2samplevsCDC51,log2samplevsCDC52,log2samplevsCDC53)
head(data2)
## GenId CDC5_1 CDC5_2 CDC5_3 pEhEx_1 pEhEx_2 pEhEx_3
## 1 EHI_000130A 45.32378 107.82734 128.66351 180.07428 446.15729 516.14317
## 2 EHI_000140A 66.04322 317.74653 257.32703 363.20067 351.59134 41.84945
## 3 EHI_000240A 700.57610 1282.45711 877.45524 1208.63415 1993.15918 2290.09468
## 4 EHI_000250A 707.05093 430.16227 388.98271 334.20566 337.04273 710.66559
## 5 EHI_000260A 94.53245 108.97444 35.15805 112.16491 69.10588 54.24928
## 6 EHI_000280A 58.27343 50.47237 80.78872 56.46397 86.07926 66.64912
## log2sample1 log2sample2 log2sample3 log2samplevsCDC51 log2samplevsCDC52
## 1 7.500438 8.804639 9.014420 5.533681 6.765897
## 2 8.508590 8.461853 5.421205 6.067020 8.316266
## 3 10.240355 10.961565 11.161821 9.454456 10.325819
## 4 8.388903 8.401062 9.475056 9.467709 8.752087
## 5 6.822283 6.131464 5.787884 6.577919 6.781024
## 6 5.844586 6.444257 6.079999 5.889314 5.685726
## log2samplevsCDC53
## 1 7.018629
## 2 8.013055
## 3 9.778825
## 4 8.607266
## 5 5.176245
## 6 6.353830
save.image('CheckPointThree.RData')
setwd("~/Documents/GitHub/Resultados/docs/PrimeroDiffExpAllResults/Clasificando/Abundances")
#load('CheckPointTwo.RData')
library(ggplot2);library(dplyr);library("fitdistrplus");
library("MASS");library("survival")
head(data2)
## GenId CDC5_1 CDC5_2 CDC5_3 pEhEx_1 pEhEx_2 pEhEx_3
## 1 EHI_000130A 45.32378 107.82734 128.66351 180.07428 446.15729 516.14317
## 2 EHI_000140A 66.04322 317.74653 257.32703 363.20067 351.59134 41.84945
## 3 EHI_000240A 700.57610 1282.45711 877.45524 1208.63415 1993.15918 2290.09468
## 4 EHI_000250A 707.05093 430.16227 388.98271 334.20566 337.04273 710.66559
## 5 EHI_000260A 94.53245 108.97444 35.15805 112.16491 69.10588 54.24928
## 6 EHI_000280A 58.27343 50.47237 80.78872 56.46397 86.07926 66.64912
## log2sample1 log2sample2 log2sample3 log2samplevsCDC51 log2samplevsCDC52
## 1 7.500438 8.804639 9.014420 5.533681 6.765897
## 2 8.508590 8.461853 5.421205 6.067020 8.316266
## 3 10.240355 10.961565 11.161821 9.454456 10.325819
## 4 8.388903 8.401062 9.475056 9.467709 8.752087
## 5 6.822283 6.131464 5.787884 6.577919 6.781024
## 6 5.844586 6.444257 6.079999 5.889314 5.685726
## log2samplevsCDC53
## 1 7.018629
## 2 8.013055
## 3 9.778825
## 4 8.607266
## 5 5.176245
## 6 6.353830
log2sample1 <- data2$log2sample1; head(mean(log2sample1)); head(sd(log2sample1))
## [1] 6.460339
## [1] 2.834041
head(log2sample1,5)
## [1] 7.500438 8.508590 10.240355 8.388903 6.822283
summary(data2)
## GenId CDC5_1 CDC5_2 CDC5_3
## Length:4691 Min. : 0.0 Min. : 0.00 Min. : 0.0
## Class :character 1st Qu.: 19.4 1st Qu.: 19.50 1st Qu.: 20.9
## Mode :character Median : 50.5 Median : 56.21 Median : 54.6
## Mean : 1930.9 Mean : 1790.28 Mean : 2024.1
## 3rd Qu.: 209.8 3rd Qu.: 248.92 3rd Qu.: 223.3
## Max. :405707.4 Max. :282737.06 Max. :384267.8
## pEhEx_1 pEhEx_2 pEhEx_3 log2sample1
## Min. : 0.00 Min. : 0.00 Min. : 0.0 Min. : 0.000
## 1st Qu.: 19.84 1st Qu.: 18.19 1st Qu.: 17.0 1st Qu.: 4.381
## Median : 55.70 Median : 56.98 Median : 60.4 Median : 5.825
## Mean : 1503.81 Mean : 1895.10 Mean : 2146.4 Mean : 6.460
## 3rd Qu.: 227.38 3rd Qu.: 250.96 3rd Qu.: 272.4 3rd Qu.: 7.835
## Max. :219640.26 Max. :288237.01 Max. :781911.9 Max. :17.745
## log2sample2 log2sample3 log2samplevsCDC51 log2samplevsCDC52
## Min. : 0.000 Min. : 0.000 Min. : 0.000 Min. : 0.000
## 1st Qu.: 4.262 1st Qu.: 4.174 1st Qu.: 4.352 1st Qu.: 4.358
## Median : 5.858 Median : 5.941 Median : 5.687 Median : 5.838
## Mean : 6.424 Mean : 6.345 Mean : 6.361 Mean : 6.456
## 3rd Qu.: 7.977 3rd Qu.: 8.095 3rd Qu.: 7.720 3rd Qu.: 7.965
## Max. :18.137 Max. :19.577 Max. :18.630 Max. :18.109
## log2samplevsCDC53
## Min. : 0.000
## 1st Qu.: 4.456
## Median : 5.797
## Mean : 6.531
## 3rd Qu.: 7.809
## Max. :18.552
ndata2 <- length(log2sample1)
hist(log2sample1, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2 sample1')
meanlog2sample1 <- mean(log2sample1); head(meanlog2sample1)
## [1] 6.460339
StdDevlog2sample1 <- sd(log2sample1); head(StdDevlog2sample1)
## [1] 2.834041
Normlog2sample1 <- (log2sample1-meanlog2sample1)/StdDevlog2sample1; head(Normlog2sample1)
## [1] 0.3670021 0.7227316 1.3337904 0.6804997 0.1277131 -0.2172704
tst<- Normlog2sample1
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2sample1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean 3.118970e-17 0.01459893
## sd 9.998934e-01 0.01032296
## Loglikelihood: -6655.741 AIC: 13315.48 BIC: 13328.39
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
Cálculo de cuantiles
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8599158 0.8169091
## 70 -0.9371527 0.9617338
## 75 -0.9960331 1.1293972
## 80 -1.0282412 1.3839029
## 85 -1.0995019 1.7041875
## 90 -1.1392596 2.1899810
## 95 -1.2814439 2.6187913
## 99 -1.6734604 3.1869302
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 BaseMean - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2sample2 <- data2$log2sample2; head(mean(log2sample2)); head(sd(log2sample2))
## [1] 6.424318
## [1] 3.078114
head(log2sample2,5)
## [1] 8.804639 8.461853 10.961565 8.401062 6.131464
ndata2 <- length(log2sample2)
hist(log2sample2, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2 sample2')
Log-normalizacion
meanlog2sample2 <- mean(log2sample2); head(meanlog2sample2)
## [1] 6.424318
StdDevlog2sample2 <- sd(log2sample2); head(StdDevlog2sample2)
## [1] 3.078114
Normlog2sample2 <- (log2sample2-meanlog2sample2)/StdDevlog2sample2; head(Normlog2sample2)
## [1] 0.773304903 0.661942766 1.474034750 0.642193202 -0.095140848
## [6] 0.006477743
tst<- Normlog2sample2
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2sample1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajuste de modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -7.845328e-17 0.01459893
## sd 9.998934e-01 0.01032296
## Loglikelihood: -6655.741 AIC: 13315.48 BIC: 13328.39
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8804743 0.8477586
## 70 -0.9259047 0.9845324
## 75 -0.9762161 1.1416416
## 80 -1.0325848 1.3729846
## 85 -1.1709305 1.7147626
## 90 -1.2592113 2.1352442
## 95 -1.5101169 2.5267006
## 99 -2.0870956 3.0982665
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 2)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 BaseMean - ADJUSTED (sample 2)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 2)', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2sample3 <- data2$log2sample3; head(mean(log2sample3)); head(sd(log2sample3))
## [1] 6.345251
## [1] 3.202711
head(log2sample3,5)
## [1] 9.014420 5.421205 11.161821 9.475056 5.787884
ndata2 <- length(log2sample3)
hist(log2sample3, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2 sample2')
meanlog2sample3 <- mean(log2sample3); head(meanlog2sample3)
## [1] 6.345251
StdDevlog2sample3 <- sd(log2sample3); head(StdDevlog2sample3)
## [1] 3.202711
Normlog2sample3 <- (log2sample3-meanlog2sample3)/StdDevlog2sample3; head(Normlog2sample3)
## [1] 0.83340927 -0.28851998 1.50390416 0.97723608 -0.17402974 -0.08282095
tst<- Normlog2sample3
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2sample1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando Modelos
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean 2.404120e-17 0.01459893
## sd 9.998934e-01 0.01032296
## Loglikelihood: -6655.741 AIC: 13315.48 BIC: 13328.39
## Correlation matrix:
## mean sd
## mean 1.000000e+00 1.713307e-11
## sd 1.713307e-11 1.000000e+00
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8675316 0.8402759
## 70 -0.9659157 0.9685687
## 75 -1.0459211 1.1104543
## 80 -1.1432722 1.3204414
## 85 -1.2676335 1.6079001
## 90 -1.4400018 1.9997962
## 95 -1.7227405 2.4584046
## 99 -1.9812123 3.0490511
Creación de histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 3)', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 3)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 BaseMean - ADJUSTED (sample 3)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 3)', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 3)', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2vsCDC51 <- data2$log2samplevsCDC51; head(mean(log2vsCDC51)); head(sd(log2vsCDC51))
## [1] 6.360533
## [1] 2.885557
head(log2vsCDC51,5)
## [1] 5.533681 6.067020 9.454456 9.467709 6.577919
ndata2 <- length(log2vsCDC51)
hist(log2vsCDC51, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2vsCDC51')
meanlog2vsCDC51 <- mean(log2vsCDC51); head(meanlog2vsCDC51)
## [1] 6.360533
StdDevlog2vsCDC51 <- sd(log2vsCDC51); head(StdDevlog2vsCDC51)
## [1] 2.885557
Normlog2vsCDC51 <- (log2vsCDC51-meanlog2vsCDC51)/StdDevlog2vsCDC51; head(Normlog2vsCDC51)
## [1] -0.28654857 -0.10171821 1.07220996 1.07680301 0.07533583 -0.16330282
tst<- Normlog2vsCDC51
Primer histograma
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2vsCDC51',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean 4.727431e-17 0.01459893
## sd 9.998934e-01 0.01032296
## Loglikelihood: -6655.741 AIC: 13315.48 BIC: 13328.39
## Correlation matrix:
## mean sd
## mean 1.000000e+00 3.426614e-11
## sd 3.426614e-11 1.000000e+00
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8422310 0.8070408
## 70 -0.8866144 0.9258394
## 75 -0.9353250 1.1082535
## 80 -0.9892988 1.3411647
## 85 -1.0498123 1.6503084
## 90 -1.1186718 2.1232202
## 95 -1.2936703 2.7259168
## 99 -2.2042655 3.3471237
Histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2vsCDC52 <- data2$log2samplevsCDC52; head(mean(log2vsCDC52)); head(sd(log2vsCDC52))
## [1] 6.456006
## [1] 2.961412
head(log2vsCDC52,5)
## [1] 6.765897 8.316266 10.325819 8.752087 6.781024
Primer Histograma
ndata2 <- length(log2vsCDC52)
hist(log2vsCDC52, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2vsCDC52')
meanlog2vsCDC52 <- mean(log2vsCDC52); head(meanlog2vsCDC52)
## [1] 6.456006
StdDevlog2vsCDC52 <- sd(log2vsCDC52); head(StdDevlog2vsCDC52)
## [1] 2.961412
Normlog2vsCDC52 <- (log2vsCDC52-meanlog2vsCDC52)/StdDevlog2vsCDC52; head(Normlog2vsCDC52)
## [1] 0.1046430 0.6281664 1.3067458 0.7753331 0.1097511 -0.2601055
tst<- Normlog2vsCDC52
** Segundo Histograma**
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2vsCDC52',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -1.572806e-17 0.01459893
## sd 9.998934e-01 0.01032296
## Loglikelihood: -6655.741 AIC: 13315.48 BIC: 13328.39
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
Cálculo de cuantiles
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8684632 0.8127421
## 70 -0.9078618 0.9668311
## 75 -0.9977360 1.1263553
## 80 -1.0497676 1.3726202
## 85 -1.1080283 1.6925494
## 90 -1.1742151 2.1332703
## 95 -1.3417825 2.5710200
## 99 -2.1800429 3.1603637
Histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2vsCDC53 <- data2$log2samplevsCDC53; head(mean(log2vsCDC53)); head(sd(log2vsCDC53))
## [1] 6.53086
## [1] 2.83016
head(log2vsCDC53,5)
## [1] 7.018629 8.013055 9.778825 8.607266 5.176245
ndata2 <- length(log2vsCDC53)
** Primer histograma**
hist(log2vsCDC53, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2vsCDC53')
meanlog2vsCDC53 <- mean(log2vsCDC53); head(meanlog2vsCDC53)
## [1] 6.53086
StdDevlog2vsCDC53 <- sd(log2vsCDC53); head(StdDevlog2vsCDC53)
## [1] 2.83016
Normlog2vsCDC53 <- (log2vsCDC53-meanlog2vsCDC53)/StdDevlog2vsCDC53; head(Normlog2vsCDC53)
## [1] 0.17234655 0.52371392 1.14762562 0.73367078 -0.47863552 -0.06255138
tst<- Normlog2vsCDC53
Segundo histograma
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2vsCDC53',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean 4.356074e-17 0.01459893
## sd 9.998934e-01 0.01032296
## Loglikelihood: -6655.741 AIC: 13315.48 BIC: 13328.39
## Correlation matrix:
## mean sd
## mean 1.000000e+00 -3.426614e-11
## sd -3.426614e-11 1.000000e+00
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
Calculo de cuantiles
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8498210 0.8013314
## 70 -0.8954944 0.9236834
## 75 -0.9456663 1.1144463
## 80 -1.0013208 1.3533967
## 85 -1.0316061 1.7113380
## 90 -1.0981755 2.2519051
## 95 -1.2178539 2.7028284
## 99 -1.5625210 3.3998336
Histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
head(pEhExvsEhMyb10,10);
## # A tibble: 10 × 7
## GenId CDC5_1 CDC5_2 CDC5_3 pEhEx_1 pEhEx_2 pEhEx_3
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 EHI_000130A 69.9 61.2 502. 140. 346. 366.
## 2 EHI_000140A 216. 28.8 65.9 281. 272. 29.6
## 3 EHI_000240A 851. 489. 12365. 936. 1544. 1622.
## 4 EHI_000250A 413. 616. 1844. 259. 261. 503.
## 5 EHI_000260A 81.6 77.4 517. 86.9 53.5 38.4
## 6 EHI_000280A 35.9 48.6 59.6 43.7 66.7 47.2
## 7 EHI_000290A 12.6 23.4 47.0 14.8 11.3 62.0
## 8 EHI_000300A 104. 68.4 9.41 106. 92.0 24.2
## 9 EHI_000410A 17.0 10.8 144. 22.5 19.7 50.0
## 10 EHI_000430A 18.8 19.8 25.1 20.7 11.3 8.24
nbreaks <- 10
data3 <- pEhExvsEhMyb10; head(data3)
## # A tibble: 6 × 7
## GenId CDC5_1 CDC5_2 CDC5_3 pEhEx_1 pEhEx_2 pEhEx_3
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 EHI_000130A 69.9 61.2 502. 140. 346. 366.
## 2 EHI_000140A 216. 28.8 65.9 281. 272. 29.6
## 3 EHI_000240A 851. 489. 12365. 936. 1544. 1622.
## 4 EHI_000250A 413. 616. 1844. 259. 261. 503.
## 5 EHI_000260A 81.6 77.4 517. 86.9 53.5 38.4
## 6 EHI_000280A 35.9 48.6 59.6 43.7 66.7 47.2
sample1 <- data3$pEhEx_1; sample2 <- data3$pEhEx_2; sample3 <- data3$pEhEx_3;
samplevs1 <- data3$CDC5_1; samplevs2 <- data3$CDC5_2; samplevs3 <- data3$CDC5_3;
log2sample1 <- log2(sample1+1); log2sample2 <- log2(sample2+1)
log2sample3 <- log2(sample3+1); log2samplevsCDC51 <- log2(samplevs1+1)
log2samplevsCDC52 <- log2(samplevs2+1); log2samplevsCDC53 <- log2(samplevs3+1)
data3 <- cbind(data3, log2sample1,log2sample2,log2sample3,
log2samplevsCDC51,log2samplevsCDC52,log2samplevsCDC53)
head(data3)
## GenId CDC5_1 CDC5_2 CDC5_3 pEhEx_1 pEhEx_2 pEhEx_3
## 1 EHI_000130A 69.92805 61.19384 501.73716 139.50869 345.55778 365.63668
## 2 EHI_000140A 216.05975 28.79710 65.85300 281.38193 272.31456 29.64622
## 3 EHI_000240A 850.79131 488.65084 12364.68511 936.36340 1543.74183 1622.30687
## 4 EHI_000250A 413.29272 616.43798 1843.88406 258.91867 261.04637 503.43668
## 5 EHI_000260A 81.58273 77.39221 517.41645 86.89736 53.52390 38.43028
## 6 EHI_000280A 35.86054 48.59511 59.58129 43.74425 66.67012 47.21435
## log2sample1 log2sample2 log2sample3 log2samplevsCDC51 log2samplevsCDC52
## 1 7.134516 8.436952 8.518207 6.148284 5.958700
## 2 8.141504 8.094418 4.937637 7.761948 4.897100
## 3 9.872465 10.593150 10.664720 9.734356 8.935610
## 4 8.021916 8.033678 8.978529 8.694507 9.270150
## 5 6.457748 5.768817 5.301232 6.367768 6.292638
## 6 5.483630 6.080447 5.591391 5.204005 5.632126
## log2samplevsCDC53
## 1 8.973661
## 2 6.062920
## 3 13.594055
## 4 10.849314
## 5 9.017968
## 6 5.920800
save.image('CheckPointFourth.RData')
setwd("~/Documents/GitHub/Resultados/docs/PrimeroDiffExpAllResults/Clasificando/Abundances")
#load('CheckPointTwo.RData')
library(ggplot2);library(dplyr);library("fitdistrplus");
library("MASS");library("survival")
head(data3)
## GenId CDC5_1 CDC5_2 CDC5_3 pEhEx_1 pEhEx_2 pEhEx_3
## 1 EHI_000130A 69.92805 61.19384 501.73716 139.50869 345.55778 365.63668
## 2 EHI_000140A 216.05975 28.79710 65.85300 281.38193 272.31456 29.64622
## 3 EHI_000240A 850.79131 488.65084 12364.68511 936.36340 1543.74183 1622.30687
## 4 EHI_000250A 413.29272 616.43798 1843.88406 258.91867 261.04637 503.43668
## 5 EHI_000260A 81.58273 77.39221 517.41645 86.89736 53.52390 38.43028
## 6 EHI_000280A 35.86054 48.59511 59.58129 43.74425 66.67012 47.21435
## log2sample1 log2sample2 log2sample3 log2samplevsCDC51 log2samplevsCDC52
## 1 7.134516 8.436952 8.518207 6.148284 5.958700
## 2 8.141504 8.094418 4.937637 7.761948 4.897100
## 3 9.872465 10.593150 10.664720 9.734356 8.935610
## 4 8.021916 8.033678 8.978529 8.694507 9.270150
## 5 6.457748 5.768817 5.301232 6.367768 6.292638
## 6 5.483630 6.080447 5.591391 5.204005 5.632126
## log2samplevsCDC53
## 1 8.973661
## 2 6.062920
## 3 13.594055
## 4 10.849314
## 5 9.017968
## 6 5.920800
log2sample1 <- data3$log2sample1; head(mean(log2sample1)); head(sd(log2sample1))
## [1] 6.089174
## [1] 2.844545
head(log2sample1,5)
## [1] 7.134516 8.141504 9.872465 8.021916 6.457748
summary(data3)
## GenId CDC5_1 CDC5_2 CDC5_3
## Length:4687 Min. : 0.00 Min. : 0.0 Min. : 0.0
## Class :character 1st Qu.: 14.34 1st Qu.: 16.2 1st Qu.: 12.5
## Mode :character Median : 41.24 Median : 41.4 Median : 50.2
## Mean : 1568.76 Mean : 1496.5 Mean : 4142.0
## 3rd Qu.: 189.61 3rd Qu.: 167.4 3rd Qu.: 239.9
## Max. :247688.75 Max. :404961.1 Max. :2325768.1
## pEhEx_1 pEhEx_2 pEhEx_3 log2sample1
## Min. : 0.00 Min. : 0.00 Min. : 0.0 Min. : 0.000
## 1st Qu.: 15.37 1st Qu.: 14.09 1st Qu.: 12.6 1st Qu.: 4.033
## Median : 43.15 Median : 44.13 Median : 43.9 Median : 5.464
## Mean : 1165.87 Mean : 1467.37 Mean : 1522.5 Mean : 6.089
## 3rd Qu.: 176.16 3rd Qu.: 194.38 3rd Qu.: 193.2 3rd Qu.: 7.469
## Max. :170161.59 Max. :223245.35 Max. :553907.7 Max. :17.377
## log2sample2 log2sample3 log2samplevsCDC51 log2samplevsCDC52
## Min. : 0.000 Min. : 0.000 Min. : 0.000 Min. : 0.000
## 1st Qu.: 3.915 1st Qu.: 3.768 1st Qu.: 3.940 1st Qu.: 4.104
## Median : 5.496 Median : 5.489 Median : 5.401 Median : 5.406
## Mean : 6.054 Mean : 5.945 Mean : 6.079 Mean : 6.086
## 3rd Qu.: 7.610 3rd Qu.: 7.602 3rd Qu.: 7.574 3rd Qu.: 7.396
## Max. :17.768 Max. :19.079 Max. :17.918 Max. :18.627
## log2samplevsCDC53
## Min. : 0.000
## 1st Qu.: 3.760
## Median : 5.677
## Mean : 6.068
## 3rd Qu.: 7.912
## Max. :21.149
ndata3 <- length(log2sample1)
hist(log2sample1, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2 sample1')
meanlog2sample1 <- mean(log2sample1); head(meanlog2sample1)
## [1] 6.089174
StdDevlog2sample1 <- sd(log2sample1); head(StdDevlog2sample1)
## [1] 2.844545
Normlog2sample1 <- (log2sample1-meanlog2sample1)/StdDevlog2sample1; head(Normlog2sample1)
## [1] 0.3674899 0.7214968 1.3300161 0.6794558 0.1295722 -0.2128788
tst<- Normlog2sample1
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2sample1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean 3.587875e-17 0.01460516
## sd 9.998933e-01 0.01032736
## Loglikelihood: -6650.065 AIC: 13304.13 BIC: 13317.03
## Correlation matrix:
## mean sd
## mean 1.000000e+00 -3.429538e-11
## sd -3.429538e-11 1.000000e+00
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
Cálculo de cuantiles
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8467216 0.8152581
## 70 -0.9222155 0.9595311
## 75 -0.9796121 1.1276277
## 80 -1.0443431 1.3803954
## 85 -1.0800958 1.7000005
## 90 -1.1601851 2.1830409
## 95 -1.3725612 2.6106296
## 99 -1.7448623 3.1766074
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 BaseMean - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2sample2 <- data3$log2sample2; head(mean(log2sample2)); head(sd(log2sample2))
## [1] 6.054099
## [1] 3.081702
head(log2sample2,5)
## [1] 8.436952 8.094418 10.593150 8.033678 5.768817
ndata3 <- length(log2sample2)
hist(log2sample2, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2 sample2')
Log-normalizacion
meanlog2sample2 <- mean(log2sample2); head(meanlog2sample2)
## [1] 6.054099
StdDevlog2sample2 <- sd(log2sample2); head(StdDevlog2sample2)
## [1] 3.081702
Normlog2sample2 <- (log2sample2-meanlog2sample2)/StdDevlog2sample2; head(Normlog2sample2)
## [1] 0.773226200 0.662075403 1.472903852 0.642365454 -0.092573117
## [6] 0.008549646
tst<- Normlog2sample2
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2sample1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajuste de modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean 1.403180e-16 0.01460516
## sd 9.998933e-01 0.01032736
## Loglikelihood: -6650.065 AIC: 13304.13 BIC: 13317.03
## Correlation matrix:
## mean sd
## mean 1.000000e+00 3.429538e-11
## sd 3.429538e-11 1.000000e+00
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8686609 0.8470080
## 70 -0.9130056 0.9828038
## 75 -0.9619944 1.1403896
## 80 -1.0786897 1.3709267
## 85 -1.1501384 1.7083438
## 90 -1.2344905 2.1268598
## 95 -1.4696481 2.5248146
## 99 -1.9645312 3.0956436
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 2)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 BaseMean - ADJUSTED (sample 2)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 2)', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2sample3 <- data3$log2sample3; head(mean(log2sample3)); head(sd(log2sample3))
## [1] 5.944632
## [1] 3.109086
head(log2sample3,5)
## [1] 8.518207 4.937637 10.664720 8.978529 5.301232
ndata3 <- length(log2sample3)
hist(log2sample3, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2 sample2')
meanlog2sample3 <- mean(log2sample3); head(meanlog2sample3)
## [1] 5.944632
StdDevlog2sample3 <- sd(log2sample3); head(StdDevlog2sample3)
## [1] 3.109086
Normlog2sample3 <- (log2sample3-meanlog2sample3)/StdDevlog2sample3; head(Normlog2sample3)
## [1] 0.8277593 -0.3238879 1.5181592 0.9758164 -0.2069419 -0.1136159
tst<- Normlog2sample3
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2sample1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando Modelos
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -1.038159e-17 0.01460516
## sd 9.998933e-01 0.01032736
## Loglikelihood: -6650.065 AIC: 13304.13 BIC: 13317.03
## Correlation matrix:
## mean sd
## mean 1.000000e+00 1.714769e-11
## sd 1.714769e-11 1.000000e+00
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8804873 0.835306
## 70 -0.9392233 0.967202
## 75 -1.0441669 1.113353
## 80 -1.1800004 1.329308
## 85 -1.2993075 1.626575
## 90 -1.3728657 2.029023
## 95 -1.5681818 2.501518
## 99 -1.9120192 3.109798
Creación de histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 3)', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 3)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 BaseMean - ADJUSTED (sample 3)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 3)', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 3)', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2vsCDC51 <- data3$log2samplevsCDC51; head(mean(log2vsCDC51)); head(sd(log2vsCDC51))
## [1] 6.079067
## [1] 3.007873
head(log2vsCDC51,5)
## [1] 6.148284 7.761948 9.734356 8.694507 6.367768
ndata3 <- length(log2vsCDC51)
hist(log2vsCDC51, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2vsCDC51')
meanlog2vsCDC51 <- mean(log2vsCDC51); head(meanlog2vsCDC51)
## [1] 6.079067
StdDevlog2vsCDC51 <- sd(log2vsCDC51); head(StdDevlog2vsCDC51)
## [1] 3.007873
Normlog2vsCDC51 <- (log2vsCDC51-meanlog2vsCDC51)/StdDevlog2vsCDC51; head(Normlog2vsCDC51)
## [1] 0.02301208 0.55949211 1.21524037 0.86953116 0.09598182 -0.29092372
tst<- Normlog2vsCDC51
Primer histograma
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2vsCDC51',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean 3.734449e-17 0.01460516
## sd 9.998933e-01 0.01032736
## Loglikelihood: -6650.065 AIC: 13304.13 BIC: 13317.03
## Correlation matrix:
## mean sd
## mean 1.000000e+00 -3.429538e-11
## sd -3.429538e-11 1.000000e+00
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8769970 0.8328201
## 70 -0.9183158 0.9594611
## 75 -0.9635325 1.1254243
## 80 -1.0134598 1.3759593
## 85 -1.0691946 1.6825955
## 90 -1.2049095 2.1281103
## 95 -1.3948811 2.6187546
## 99 -2.0210515 3.1616174
Histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2vsCDC52 <- data3$log2samplevsCDC52; head(mean(log2vsCDC52)); head(sd(log2vsCDC52))
## [1] 6.08571
## [1] 2.815509
head(log2vsCDC52,5)
## [1] 5.958700 4.897100 8.935610 9.270150 6.292638
Primer Histograma
ndata3 <- length(log2vsCDC52)
hist(log2vsCDC52, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2vsCDC52')
meanlog2vsCDC52 <- mean(log2vsCDC52); head(meanlog2vsCDC52)
## [1] 6.08571
StdDevlog2vsCDC52 <- sd(log2vsCDC52); head(StdDevlog2vsCDC52)
## [1] 2.815509
Normlog2vsCDC52 <- (log2vsCDC52-meanlog2vsCDC52)/StdDevlog2vsCDC52; head(Normlog2vsCDC52)
## [1] -0.04511077 -0.42216506 1.01221485 1.13103560 0.07349607 -0.16110183
tst<- Normlog2vsCDC52
** Segundo Histograma**
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2vsCDC52',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean 7.844575e-18 0.01460516
## sd 9.998933e-01 0.01032736
## Loglikelihood: -6650.065 AIC: 13304.13 BIC: 13317.03
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
Cálculo de cuantiles
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8241172 0.7977969
## 70 -0.8968636 0.9352333
## 75 -0.9375162 1.0847564
## 80 -0.9816741 1.3463833
## 85 -1.0299993 1.6629255
## 90 -1.0833612 2.1783677
## 95 -1.2880076 2.7298034
## 99 -2.1614954 3.4267131
Histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2vsCDC53 <- data3$log2samplevsCDC53; head(mean(log2vsCDC53)); head(sd(log2vsCDC53))
## [1] 6.068265
## [1] 3.43608
head(log2vsCDC53,5)
## [1] 8.973661 6.062920 13.594055 10.849314 9.017968
ndata3 <- length(log2vsCDC53)
** Primer histograma**
hist(log2vsCDC53, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2vsCDC53')
meanlog2vsCDC53 <- mean(log2vsCDC53); head(meanlog2vsCDC53)
## [1] 6.068265
StdDevlog2vsCDC53 <- sd(log2vsCDC53); head(StdDevlog2vsCDC53)
## [1] 3.43608
Normlog2vsCDC53 <- (log2vsCDC53-meanlog2vsCDC53)/StdDevlog2vsCDC53; head(Normlog2vsCDC53)
## [1] 0.845555235 -0.001555479 2.190225383 1.391425525 0.858449916
## [6] -0.042916597
tst<- Normlog2vsCDC53
Segundo histograma
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2vsCDC53',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -1.107054e-16 0.01460516
## sd 9.998933e-01 0.01032736
## Loglikelihood: -6650.065 AIC: 13304.13 BIC: 13317.03
## Correlation matrix:
## mean sd
## mean 1.000000e+00 1.714769e-11
## sd 1.714769e-11 1.000000e+00
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
Calculo de cuantiles
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.7824916 0.8709604
## 70 -0.9330312 0.9884576
## 75 -1.1699608 1.1455636
## 80 -1.1699608 1.3353109
## 85 -1.1699608 1.5512574
## 90 -1.7660431 1.8682788
## 95 -1.7660431 2.3078117
## 99 -1.7660431 3.1171283
Histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
head(pEhExvsU2AF84,10);
## # A tibble: 10 × 7
## GenId CDC5_1 CDC5_2 CDC5_3 pEhEx_1 pEhEx_2 pEhEx_3
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 EHI_000130A 59.6 70.4 611. 140. 347. 382.
## 2 EHI_000140A 118. 207. 91.9 283. 273. 31.0
## 3 EHI_000240A 689. 871. 11090. 941. 1550. 1694.
## 4 EHI_000250A 426. 407. 1608. 260. 262. 526.
## 5 EHI_000260A 104. 110. 422. 87.3 53.7 40.1
## 6 EHI_000280A 40.4 35.2 108. 43.9 66.9 49.3
## 7 EHI_000290A 15.2 19.7 44.8 14.8 11.3 64.8
## 8 EHI_000300A 74.8 94.2 2.36 107. 92.4 25.2
## 9 EHI_000410A 15.2 15.5 134. 22.6 19.8 52.2
## 10 EHI_000430A 18.2 23.8 28.3 20.8 11.3 8.60
nbreaks <- 10
data4 <- pEhExvsU2AF84; head(data4)
## # A tibble: 6 × 7
## GenId CDC5_1 CDC5_2 CDC5_3 pEhEx_1 pEhEx_2 pEhEx_3
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 EHI_000130A 59.6 70.4 611. 140. 347. 382.
## 2 EHI_000140A 118. 207. 91.9 283. 273. 31.0
## 3 EHI_000240A 689. 871. 11090. 941. 1550. 1694.
## 4 EHI_000250A 426. 407. 1608. 260. 262. 526.
## 5 EHI_000260A 104. 110. 422. 87.3 53.7 40.1
## 6 EHI_000280A 40.4 35.2 108. 43.9 66.9 49.3
sample1 <- data4$pEhEx_1; sample2 <- data4$pEhEx_2; sample3 <- data4$pEhEx_3;
samplevs1 <- data4$CDC5_1; samplevs2 <- data4$CDC5_2; samplevs3 <- data4$CDC5_3;
log2sample1 <- log2(sample1+1); log2sample2 <- log2(sample2+1)
log2sample3 <- log2(sample3+1); log2samplevsCDC51 <- log2(samplevs1+1)
log2samplevsCDC52 <- log2(samplevs2+1); log2samplevsCDC53 <- log2(samplevs3+1)
data4 <- cbind(data4, log2sample1,log2sample2,log2sample3,
log2samplevsCDC51,log2samplevsCDC52,log2samplevsCDC53)
head(data4)
## GenId CDC5_1 CDC5_2 CDC5_3 pEhEx_1 pEhEx_2 pEhEx_3
## 1 EHI_000130A 59.61996 70.37838 610.61850 140.16143 346.92628 381.74940
## 2 EHI_000140A 118.22941 206.99523 91.94642 282.69848 273.39300 30.95265
## 3 EHI_000240A 689.16628 871.44992 11090.15226 940.74452 1549.85546 1693.79799
## 4 EHI_000250A 426.43427 406.74563 1607.88347 260.13011 262.08018 525.62192
## 5 EHI_000260A 104.08230 109.70747 422.01047 87.30394 53.73586 40.12381
## 6 EHI_000280A 40.42031 35.18919 108.44962 43.94892 66.93415 49.29497
## log2sample1 log2sample2 log2sample3 log2samplevsCDC51 log2samplevsCDC52
## 1 7.141202 8.442638 8.580256 5.921721 6.157415
## 2 8.148215 8.100100 4.997864 6.897596 7.700407
## 3 9.879192 10.598849 10.726898 9.430800 9.768929
## 4 8.028625 8.039359 9.040624 8.739559 8.671526
## 5 6.464406 5.774415 5.361902 6.715376 6.790609
## 6 5.490215 6.086065 5.652342 5.372266 5.177487
## log2samplevsCDC53
## 1 9.256488
## 2 6.538327
## 3 13.437122
## 4 10.651844
## 5 8.724550
## 6 6.774123
save.image('CheckPointFifth.RData')
setwd("~/Documents/GitHub/Resultados/docs/PrimeroDiffExpAllResults/Clasificando/Abundances")
#load('CheckPointTwo.RData')
library(ggplot2);library(dplyr);library("fitdistrplus");
library("MASS");library("survival")
head(data4)
## GenId CDC5_1 CDC5_2 CDC5_3 pEhEx_1 pEhEx_2 pEhEx_3
## 1 EHI_000130A 59.61996 70.37838 610.61850 140.16143 346.92628 381.74940
## 2 EHI_000140A 118.22941 206.99523 91.94642 282.69848 273.39300 30.95265
## 3 EHI_000240A 689.16628 871.44992 11090.15226 940.74452 1549.85546 1693.79799
## 4 EHI_000250A 426.43427 406.74563 1607.88347 260.13011 262.08018 525.62192
## 5 EHI_000260A 104.08230 109.70747 422.01047 87.30394 53.73586 40.12381
## 6 EHI_000280A 40.42031 35.18919 108.44962 43.94892 66.93415 49.29497
## log2sample1 log2sample2 log2sample3 log2samplevsCDC51 log2samplevsCDC52
## 1 7.141202 8.442638 8.580256 5.921721 6.157415
## 2 8.148215 8.100100 4.997864 6.897596 7.700407
## 3 9.879192 10.598849 10.726898 9.430800 9.768929
## 4 8.028625 8.039359 9.040624 8.739559 8.671526
## 5 6.464406 5.774415 5.361902 6.715376 6.790609
## 6 5.490215 6.086065 5.652342 5.372266 5.177487
## log2samplevsCDC53
## 1 9.256488
## 2 6.538327
## 3 13.437122
## 4 10.651844
## 5 8.724550
## 6 6.774123
log2sample1 <- data4$log2sample1; head(mean(log2sample1)); head(sd(log2sample1))
## [1] 6.066239
## [1] 2.837457
head(log2sample1,5)
## [1] 7.141202 8.148215 9.879192 8.028625 6.464406
summary(data4)
## GenId CDC5_1 CDC5_2 CDC5_3
## Length:4746 Min. : 0.00 Min. : 0.00 Min. : 0.0
## Class :character 1st Qu.: 15.16 1st Qu.: 14.49 1st Qu.: 14.1
## Mode :character Median : 41.43 Median : 42.43 Median : 49.5
## Mean : 1422.93 Mean : 1431.84 Mean : 3498.7
## 3rd Qu.: 168.75 3rd Qu.: 187.33 3rd Qu.: 237.5
## Max. :236529.55 Max. :222485.72 Max. :1942165.3
## pEhEx_1 pEhEx_2 pEhEx_3 log2sample1
## Min. : 0.00 Min. : 0.00 Min. : 0.0 Min. : 0.000
## 1st Qu.: 14.85 1st Qu.: 13.20 1st Qu.: 12.6 1st Qu.: 3.986
## Median : 42.17 Median : 42.42 Median : 44.7 Median : 5.432
## Mean : 1156.92 Mean : 1454.96 Mean : 1569.6 Mean : 6.066
## 3rd Qu.: 172.68 3rd Qu.: 191.38 3rd Qu.: 199.5 3rd Qu.: 7.440
## Max. :170957.75 Max. :224129.46 Max. :578317.1 Max. :17.383
## log2sample2 log2sample3 log2samplevsCDC51 log2samplevsCDC52
## Min. : 0.000 Min. : 0.000 Min. : 0.000 Min. : 0.000
## 1st Qu.: 3.828 1st Qu.: 3.767 1st Qu.: 4.014 1st Qu.: 3.953
## Median : 5.440 Median : 5.514 Median : 5.407 Median : 5.441
## Mean : 6.019 Mean : 5.939 Mean : 6.074 Mean : 6.080
## 3rd Qu.: 7.588 3rd Qu.: 7.647 3rd Qu.: 7.407 3rd Qu.: 7.557
## Max. :17.774 Max. :19.142 Max. :17.852 Max. :17.763
## log2samplevsCDC53
## Min. : 0.000
## 1st Qu.: 3.921
## Median : 5.658
## Mean : 6.152
## 3rd Qu.: 7.898
## Max. :20.889
ndata4 <- length(log2sample1)
hist(log2sample1, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2 sample1')
meanlog2sample1 <- mean(log2sample1); head(meanlog2sample1)
## [1] 6.066239
StdDevlog2sample1 <- sd(log2sample1); head(StdDevlog2sample1)
## [1] 2.837457
Normlog2sample1 <- (log2sample1-meanlog2sample1)/StdDevlog2sample1; head(Normlog2sample1)
## [1] 0.3788472 0.7337468 1.3437920 0.6916001 0.1403251 -0.2030074
tst<- Normlog2sample1
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2sample1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -5.346355e-17 0.01451411
## sd 9.998946e-01 0.01026298
## Loglikelihood: -6733.782 AIC: 13471.56 BIC: 13484.49
## Correlation matrix:
## mean sd
## mean 1.000000e+00 -3.386913e-11
## sd -3.386913e-11 1.000000e+00
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
Cálculo de cuantiles
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8625714 0.8124762
## 70 -0.9422460 0.9640896
## 75 -0.9718431 1.1342860
## 80 -1.0367684 1.3790909
## 85 -1.0726303 1.7033941
## 90 -1.1529698 2.1934547
## 95 -1.3038487 2.6233257
## 99 -1.7398510 3.1888833
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 BaseMean - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2sample2 <- data4$log2sample2; head(mean(log2sample2)); head(sd(log2sample2))
## [1] 6.018835
## [1] 3.085311
head(log2sample2,5)
## [1] 8.442638 8.100100 10.598849 8.039359 5.774415
ndata4 <- length(log2sample2)
hist(log2sample2, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2 sample2')
Log-normalizacion
meanlog2sample2 <- mean(log2sample2); head(meanlog2sample2)
## [1] 6.018835
StdDevlog2sample2 <- sd(log2sample2); head(StdDevlog2sample2)
## [1] 3.085311
Normlog2sample2 <- (log2sample2-meanlog2sample2)/StdDevlog2sample2; head(Normlog2sample2)
## [1] 0.78559421 0.67457202 1.48445747 0.65488484 -0.07922075 0.02179028
tst<- Normlog2sample2
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2sample1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajuste de modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean 4.519200e-17 0.01451411
## sd 9.998946e-01 0.01026298
## Loglikelihood: -6733.782 AIC: 13471.56 BIC: 13484.49
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8988548 0.8463966
## 70 -0.9478079 0.9891035
## 75 -1.0024917 1.1467315
## 80 -1.0644280 1.3749391
## 85 -1.1358390 1.6937524
## 90 -1.2201564 2.1241067
## 95 -1.4552922 2.5236980
## 99 -1.9508033 3.0984441
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 2)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 BaseMean - ADJUSTED (sample 2)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 2)', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2sample3 <- data4$log2sample3; head(mean(log2sample3)); head(sd(log2sample3))
## [1] 5.938851
## [1] 3.137257
head(log2sample3,5)
## [1] 8.580256 4.997864 10.726898 9.040624 5.361902
ndata4 <- length(log2sample3)
hist(log2sample3, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2 sample2')
meanlog2sample3 <- mean(log2sample3); head(meanlog2sample3)
## [1] 5.938851
StdDevlog2sample3 <- sd(log2sample3); head(StdDevlog2sample3)
## [1] 3.137257
Normlog2sample3 <- (log2sample3-meanlog2sample3)/StdDevlog2sample3; head(Normlog2sample3)
## [1] 0.84194733 -0.29993956 1.52618884 0.98868937 -0.18390246 -0.09132474
tst<- Normlog2sample3
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2sample1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando Modelos
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean 1.897633e-17 0.01451411
## sd 9.998946e-01 0.01026298
## Loglikelihood: -6733.782 AIC: 13471.56 BIC: 13484.49
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8813312 0.8377086
## 70 -0.9785407 0.9709584
## 75 -1.0570406 1.1167404
## 80 -1.1517563 1.3292508
## 85 -1.2711789 1.6268787
## 90 -1.4329274 2.0299674
## 95 -1.5417724 2.4960043
## 99 -1.8930077 3.1029256
Creación de histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 3)', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 3)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 BaseMean - ADJUSTED (sample 3)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 3)', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 3)', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2vsCDC51 <- data4$log2samplevsCDC51; head(mean(log2vsCDC51)); head(sd(log2vsCDC51))
## [1] 6.073889
## [1] 2.856279
head(log2vsCDC51,5)
## [1] 5.921721 6.897596 9.430800 8.739559 6.715376
ndata4 <- length(log2vsCDC51)
hist(log2vsCDC51, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2vsCDC51')
meanlog2vsCDC51 <- mean(log2vsCDC51); head(meanlog2vsCDC51)
## [1] 6.073889
StdDevlog2vsCDC51 <- sd(log2vsCDC51); head(StdDevlog2vsCDC51)
## [1] 2.856279
Normlog2vsCDC51 <- (log2vsCDC51-meanlog2vsCDC51)/StdDevlog2vsCDC51; head(Normlog2vsCDC51)
## [1] -0.05327506 0.28838461 1.17527408 0.93326643 0.22458816 -0.24564228
tst<- Normlog2vsCDC51
Primer histograma
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2vsCDC51',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -1.363897e-18 0.01451411
## sd 9.998946e-01 0.01026298
## Loglikelihood: -6733.782 AIC: 13471.56 BIC: 13484.49
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8665460 0.8005323
## 70 -0.9105349 0.9400738
## 75 -0.9587235 1.1113009
## 80 -1.0119991 1.3461767
## 85 -1.0715638 1.6701383
## 90 -1.0715638 2.1555034
## 95 -1.3093554 2.7085444
## 99 -2.1265041 3.2907409
Histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2vsCDC52 <- data4$log2samplevsCDC52; head(mean(log2vsCDC52)); head(sd(log2vsCDC52))
## [1] 6.079551
## [1] 2.958446
head(log2vsCDC52,5)
## [1] 6.157415 7.700407 9.768929 8.671526 6.790609
Primer Histograma
ndata4 <- length(log2vsCDC52)
hist(log2vsCDC52, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2vsCDC52')
meanlog2vsCDC52 <- mean(log2vsCDC52); head(meanlog2vsCDC52)
## [1] 6.079551
StdDevlog2vsCDC52 <- sd(log2vsCDC52); head(StdDevlog2vsCDC52)
## [1] 2.958446
Normlog2vsCDC52 <- (log2vsCDC52-meanlog2vsCDC52)/StdDevlog2vsCDC52; head(Normlog2vsCDC52)
## [1] 0.02631943 0.54787417 1.24706626 0.87612726 0.24034856 -0.30491137
tst<- Normlog2vsCDC52
** Segundo Histograma**
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2vsCDC52',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -9.044811e-17 0.01451411
## sd 9.998946e-01 0.01026298
## Loglikelihood: -6733.782 AIC: 13471.56 BIC: 13484.49
## Correlation matrix:
## mean sd
## mean 1.000000e+00 -3.386913e-11
## sd -3.386913e-11 1.000000e+00
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
Cálculo de cuantiles
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8703761 0.8136491
## 70 -0.9170047 0.9590243
## 75 -0.9685679 1.1169792
## 80 -1.0262349 1.3717093
## 85 -1.0916475 1.6775570
## 90 -1.1672140 2.1543218
## 95 -1.3663234 2.6103392
## 99 -2.0549813 3.1594798
Histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC52 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2vsCDC53 <- data4$log2samplevsCDC53; head(mean(log2vsCDC53)); head(sd(log2vsCDC53))
## [1] 6.151641
## [1] 3.315575
head(log2vsCDC53,5)
## [1] 9.256488 6.538327 13.437122 10.651844 8.724550
ndata4 <- length(log2vsCDC53)
** Primer histograma**
hist(log2vsCDC53, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2vsCDC53')
meanlog2vsCDC53 <- mean(log2vsCDC53); head(meanlog2vsCDC53)
## [1] 6.151641
StdDevlog2vsCDC53 <- sd(log2vsCDC53); head(StdDevlog2vsCDC53)
## [1] 3.315575
Normlog2vsCDC53 <- (log2vsCDC53-meanlog2vsCDC53)/StdDevlog2vsCDC53; head(Normlog2vsCDC53)
## [1] 0.9364429 0.1166271 2.1973504 1.3572917 0.7760066 0.1877447
tst<- Normlog2vsCDC53
Segundo histograma
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2vsCDC53',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -1.080389e-16 0.01451411
## sd 9.998946e-01 0.01026298
## Loglikelihood: -6733.782 AIC: 13471.56 BIC: 13484.49
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
Calculo de cuantiles
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8351247 0.8675257
## 70 -0.9466147 0.9977469
## 75 -0.9466147 1.1516663
## 80 -1.0968948 1.3476203
## 85 -1.3283399 1.5933747
## 90 -1.3283399 1.8884142
## 95 -1.8553769 2.3619041
## 99 -1.8553769 3.1601836
Histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC51 - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsCDC53 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
head(pEhExvsUmasM,10);
## # A tibble: 10 × 7
## GenId UmasM_1 UmasM_2 UmasM_3 pEhEx_1 pEhEx_2 pEhEx_3
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 EHI_000130A 52.8 428. 877. 175. 432. 478.
## 2 EHI_000140A 118. 371. 43.7 353. 341. 38.8
## 3 EHI_000240A 744. 1165. 7104. 1175. 1932. 2122.
## 4 EHI_000250A 704. 213. 1578. 325. 327. 659.
## 5 EHI_000260A 119. 96.2 139. 109. 67.0 50.3
## 6 EHI_000280A 57.6 59.6 34.3 54.9 83.4 61.8
## 7 EHI_000290A 26.4 19.7 70.2 18.5 14.1 81.2
## 8 EHI_000300A 80.4 115. 15.6 134. 115. 31.6
## 9 EHI_000410A 20.4 24.4 139. 28.2 24.7 65.4
## 10 EHI_000430A 31.2 35.9 14.0 26.0 14.1 10.8
nbreaks <- 10
data5 <- pEhExvsUmasM; head(data5)
## # A tibble: 6 × 7
## GenId UmasM_1 UmasM_2 UmasM_3 pEhEx_1 pEhEx_2 pEhEx_3
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 EHI_000130A 52.8 428. 877. 175. 432. 478.
## 2 EHI_000140A 118. 371. 43.7 353. 341. 38.8
## 3 EHI_000240A 744. 1165. 7104. 1175. 1932. 2122.
## 4 EHI_000250A 704. 213. 1578. 325. 327. 659.
## 5 EHI_000260A 119. 96.2 139. 109. 67.0 50.3
## 6 EHI_000280A 57.6 59.6 34.3 54.9 83.4 61.8
sample1 <- data5$pEhEx_1; sample2 <- data5$pEhEx_2; sample3 <- data5$pEhEx_3;
samplevs1 <- data5$UmasM_1; samplevs2 <- data5$UmasM_2; samplevs3 <- data5$UmasM_3;
log2sample1 <- log2(sample1+1); log2sample2 <- log2(sample2+1)
log2sample3 <- log2(sample3+1); log2samplevsumasM1 <- log2(samplevs1+1)
log2samplevsumasM2 <- log2(samplevs2+1); log2samplevsumasM3 <- log2(samplevs3+1)
data5 <- cbind(data5, log2sample1,log2sample2,log2sample3,
log2samplevsumasM1,log2samplevsumasM2,log2samplevsumasM3)
head(data5)
## GenId UmasM_1 UmasM_2 UmasM_3 pEhEx_1 pEhEx_2 pEhEx_3
## 1 EHI_000130A 52.79569 428.31302 877.32975 175.05490 432.49403 478.29632
## 2 EHI_000140A 117.59041 370.70763 43.71038 353.07682 340.82410 38.78078
## 3 EHI_000240A 743.93931 1164.98430 7104.49766 1174.94473 1932.12006 2122.17063
## 4 EHI_000250A 704.34254 212.80109 1578.25690 324.89002 326.72103 658.55515
## 5 EHI_000260A 118.79031 96.23489 138.93656 109.03843 66.98956 50.27139
## 6 EHI_000280A 57.59530 59.63852 34.34387 54.89009 83.44314 61.76199
## log2sample1 log2sample2 log2sample3 log2samplevsumasM1 log2samplevsumasM2
## 1 7.459882 8.759868 8.904774 5.749419 8.745886
## 2 8.467919 8.417110 5.314000 6.889844 8.538024
## 3 10.199605 10.916716 11.052005 9.540979 10.187333
## 4 8.348241 8.356324 9.365349 9.462180 7.740125
## 5 6.781864 6.087241 5.680082 6.904367 6.603402
## 6 5.804521 6.399908 5.971819 5.872713 5.922163
## log2samplevsumasM3
## 1 9.778619
## 2 5.482538
## 3 12.794720
## 4 10.625030
## 5 7.128629
## 6 5.143388
save.image('CheckPointSixth.RData')
setwd("~/Documents/GitHub/Resultados/docs/PrimeroDiffExpAllResults/Clasificando/Abundances")
#load('CheckPointTwo.RData')
library(ggplot2);library(dplyr);library("fitdistrplus");
library("MASS");library("survival")
head(data5)
## GenId UmasM_1 UmasM_2 UmasM_3 pEhEx_1 pEhEx_2 pEhEx_3
## 1 EHI_000130A 52.79569 428.31302 877.32975 175.05490 432.49403 478.29632
## 2 EHI_000140A 117.59041 370.70763 43.71038 353.07682 340.82410 38.78078
## 3 EHI_000240A 743.93931 1164.98430 7104.49766 1174.94473 1932.12006 2122.17063
## 4 EHI_000250A 704.34254 212.80109 1578.25690 324.89002 326.72103 658.55515
## 5 EHI_000260A 118.79031 96.23489 138.93656 109.03843 66.98956 50.27139
## 6 EHI_000280A 57.59530 59.63852 34.34387 54.89009 83.44314 61.76199
## log2sample1 log2sample2 log2sample3 log2samplevsumasM1 log2samplevsumasM2
## 1 7.459882 8.759868 8.904774 5.749419 8.745886
## 2 8.467919 8.417110 5.314000 6.889844 8.538024
## 3 10.199605 10.916716 11.052005 9.540979 10.187333
## 4 8.348241 8.356324 9.365349 9.462180 7.740125
## 5 6.781864 6.087241 5.680082 6.904367 6.603402
## 6 5.804521 6.399908 5.971819 5.872713 5.922163
## log2samplevsumasM3
## 1 9.778619
## 2 5.482538
## 3 12.794720
## 4 10.625030
## 5 7.128629
## 6 5.143388
log2sample1 <- data5$log2sample1; head(mean(log2sample1)); head(sd(log2sample1))
## [1] 6.24652
## [1] 2.884429
head(log2sample1,5)
## [1] 7.459882 8.467919 10.199605 8.348241 6.781864
summary(data5)
## GenId UmasM_1 UmasM_2 UmasM_3
## Length:4919 Min. : 0.0 Min. : 0.00 Min. : 0.0
## Class :character 1st Qu.: 16.8 1st Qu.: 18.30 1st Qu.: 14.0
## Mode :character Median : 46.8 Median : 52.18 Median : 54.6
## Mean : 1915.1 Mean : 1003.67 Mean : 3444.8
## 3rd Qu.: 198.0 3rd Qu.: 193.15 3rd Qu.: 295.0
## Max. :340994.2 Max. :145896.83 Max. :1488475.8
## pEhEx_1 pEhEx_2 pEhEx_3 log2sample1
## Min. : 0.00 Min. : 0.00 Min. : 0.0 Min. : 0.000
## 1st Qu.: 16.32 1st Qu.: 15.28 1st Qu.: 15.1 1st Qu.: 4.114
## Median : 48.96 Median : 48.19 Median : 52.4 Median : 5.643
## Mean : 1394.44 Mean : 1752.18 Mean : 1898.2 Mean : 6.247
## 3rd Qu.: 199.53 3rd Qu.: 222.71 3rd Qu.: 231.2 3rd Qu.: 7.648
## Max. :213518.02 Max. :279409.95 Max. :724577.3 Max. :17.704
## log2sample2 log2sample3 log2samplevsumasM1 log2samplevsumasM2
## Min. : 0.000 Min. : 0.000 Min. : 0.000 Min. : 0.000
## 1st Qu.: 4.025 1st Qu.: 4.007 1st Qu.: 4.154 1st Qu.: 4.270
## Median : 5.620 Median : 5.739 Median : 5.579 Median : 5.733
## Mean : 6.187 Mean : 6.152 Mean : 6.222 Mean : 6.256
## 3rd Qu.: 7.805 3rd Qu.: 7.860 3rd Qu.: 7.637 3rd Qu.: 7.601
## Max. :18.092 Max. :19.467 Max. :18.379 Max. :17.155
## log2samplevsumasM3
## Min. : 0.000
## 1st Qu.: 3.912
## Median : 5.798
## Mean : 6.317
## 3rd Qu.: 8.210
## Max. :20.505
ndata5 <- length(log2sample1)
hist(log2sample1, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2 sample1')
meanlog2sample1 <- mean(log2sample1); head(meanlog2sample1)
## [1] 6.24652
StdDevlog2sample1 <- sd(log2sample1); head(StdDevlog2sample1)
## [1] 2.884429
Normlog2sample1 <- (log2sample1-meanlog2sample1)/StdDevlog2sample1; head(Normlog2sample1)
## [1] 0.4206592 0.7701346 1.3704911 0.7286438 0.1855978 -0.1532363
tst<- Normlog2sample1
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2sample1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -1.310539e-16 0.01425665
## sd 9.998983e-01 0.01008093
## Loglikelihood: -6979.259 AIC: 13962.52 BIC: 13975.52
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
Cálculo de cuantiles
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8877976 0.8234548
## 70 -0.9174931 0.9668385
## 75 -0.9827617 1.1346553
## 80 -1.0188953 1.3615138
## 85 -1.1000839 1.6923518
## 90 -1.1462228 2.1593960
## 95 -1.3906263 2.6069712
## 99 -2.1655999 3.1547856
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 BaseMean - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2sample2 <- data5$log2sample2; head(mean(log2sample2)); head(sd(log2sample2))
## [1] 6.18657
## [1] 3.14197
head(log2sample2,5)
## [1] 8.759868 8.417110 10.916716 8.356324 6.087241
ndata5 <- length(log2sample2)
hist(log2sample2, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2 sample2')
Log-normalizacion
meanlog2sample2 <- mean(log2sample2); head(meanlog2sample2)
## [1] 6.18657
StdDevlog2sample2 <- sd(log2sample2); head(StdDevlog2sample2)
## [1] 3.14197
Normlog2sample2 <- (log2sample2-meanlog2sample2)/StdDevlog2sample2; head(Normlog2sample2)
## [1] 0.81900802 0.70991786 1.50547137 0.69057145 -0.03161338 0.06789964
tst<- Normlog2sample2
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2sample1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajuste de modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -1.093861e-16 0.01425665
## sd 9.998983e-01 0.01008093
## Loglikelihood: -6979.259 AIC: 13962.52 BIC: 13975.52
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8936351 0.8467827
## 70 -0.8936351 0.9884290
## 75 -1.0112465 1.1344069
## 80 -1.0836965 1.3626690
## 85 -1.1697595 1.6777371
## 90 -1.2757625 2.1128235
## 95 -1.6121684 2.5115181
## 99 -1.9690098 3.0656774
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 2)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 BaseMean - ADJUSTED (sample 2)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 2)', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2sample3 <- data5$log2sample3; head(mean(log2sample3)); head(sd(log2sample3))
## [1] 6.152478
## [1] 3.158081
head(log2sample3,5)
## [1] 8.904774 5.314000 11.052005 9.365349 5.680082
ndata5 <- length(log2sample3)
hist(log2sample3, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2 sample2')
meanlog2sample3 <- mean(log2sample3); head(meanlog2sample3)
## [1] 6.152478
StdDevlog2sample3 <- sd(log2sample3); head(StdDevlog2sample3)
## [1] 3.158081
Normlog2sample3 <- (log2sample3-meanlog2sample3)/StdDevlog2sample3; head(Normlog2sample3)
## [1] 0.87150918 -0.26550237 1.55142547 1.01734948 -0.14958317 -0.05720518
tst<- Normlog2sample3
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2sample1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando Modelos
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean 1.292160e-16 0.01425665
## sd 9.998983e-01 0.01008093
## Loglikelihood: -6979.259 AIC: 13962.52 BIC: 13975.52
## Correlation matrix:
## mean sd
## mean 1.00000e+00 -3.26782e-11
## sd -3.26782e-11 1.00000e+00
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8811857 0.8359346
## 70 -0.9495345 0.9646852
## 75 -1.0299396 1.1073525
## 80 -1.1855441 1.3225658
## 85 -1.2519401 1.6081997
## 90 -1.4233556 2.0263238
## 95 -1.7009100 2.4893700
## 99 -1.9481699 3.1039165
Creación de histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 3)', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 3)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 2)', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 BaseMean - ADJUSTED (sample 3)', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - DATA (sample 3)', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2 pEhEx - ADJUSTED (sample 3)', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2vsumasM1 <- data5$log2samplevsumasM1; head(mean(log2vsumasM1)); head(sd(log2vsumasM1))
## [1] 6.221901
## [1] 2.955459
head(log2vsumasM1,5)
## [1] 5.749419 6.889844 9.540979 9.462180 6.904367
ndata5 <- length(log2vsumasM1)
hist(log2vsumasM1, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2vsumasM1')
meanlog2vsumasM1 <- mean(log2vsumasM1); head(meanlog2vsumasM1)
## [1] 6.221901
StdDevlog2vsumasM1 <- sd(log2vsumasM1); head(StdDevlog2vsumasM1)
## [1] 2.955459
Normlog2vsumasM1 <- (log2vsumasM1-meanlog2vsumasM1)/StdDevlog2vsumasM1; head(Normlog2vsumasM1)
## [1] -0.1598676 0.2260030 1.1230332 1.0963710 0.2309173 -0.1181501
tst<- Normlog2vsumasM1
Primer histograma
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2vsumasM1',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -9.853035e-17 0.01425665
## sd 9.998983e-01 0.01008093
## Loglikelihood: -6979.259 AIC: 13962.52 BIC: 13975.52
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8531906 0.8236011
## 70 -0.9004672 0.9332295
## 75 -0.9528182 1.1118993
## 80 -0.9528182 1.3280269
## 85 -1.0781339 1.6545578
## 90 -1.1553694 2.1045904
## 95 -1.3603164 2.6974724
## 99 -2.1052233 3.3025359
Histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM1 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM1 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM1 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM1 - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM1 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM1 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2vsumasM2 <- data5$log2samplevsumasM2; head(mean(log2vsumasM2)); head(sd(log2vsumasM2))
## [1] 6.255602
## [1] 2.72471
head(log2vsumasM2,5)
## [1] 8.745886 8.538024 10.187333 7.740125 6.603402
Primer Histograma
ndata5 <- length(log2vsumasM2)
hist(log2vsumasM2, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2vsumasM2')
meanlog2vsumasM2 <- mean(log2vsumasM2); head(meanlog2vsumasM2)
## [1] 6.255602
StdDevlog2vsumasM2 <- sd(log2vsumasM2); head(StdDevlog2vsumasM2)
## [1] 2.72471
Normlog2vsumasM2 <- (log2vsumasM2-meanlog2vsumasM2)/StdDevlog2vsumasM2; head(Normlog2vsumasM2)
## [1] 0.9139630 0.8376754 1.4429905 0.5448372 0.1276466 -0.1223761
tst<- Normlog2vsumasM2
** Segundo Histograma**
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2vsumasM2',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean 1.626526e-16 0.01425665
## sd 9.998983e-01 0.01008093
## Loglikelihood: -6979.259 AIC: 13962.52 BIC: 13975.52
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
Cálculo de cuantiles
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8779788 0.8200071
## 70 -0.9297388 0.9699103
## 75 -0.9871123 1.1247423
## 80 -1.0514667 1.3424300
## 85 -1.1247392 1.7188713
## 90 -1.2098054 2.1841922
## 95 -1.3112040 2.5940822
## 99 -1.8422575 3.1504680
Histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM2 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM2 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM2 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM2 - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM2 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM2 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
log2vsumasM3 <- data5$log2samplevsumasM3; head(mean(log2vsumasM3)); head(sd(log2vsumasM3))
## [1] 6.317092
## [1] 3.410882
head(log2vsumasM3,5)
## [1] 9.778619 5.482538 12.794720 10.625030 7.128629
ndata5 <- length(log2vsumasM3)
** Primer histograma**
hist(log2vsumasM3, breaks = nbreaks, col= rainbow(25,0.3),
main = 'Log2vsumasM3')
meanlog2vsumasM3 <- mean(log2vsumasM3); head(meanlog2vsumasM3)
## [1] 6.317092
StdDevlog2vsumasM3 <- sd(log2vsumasM3); head(StdDevlog2vsumasM3)
## [1] 3.410882
Normlog2vsumasM3 <- (log2vsumasM3-meanlog2vsumasM3)/StdDevlog2vsumasM3; head(Normlog2vsumasM3)
## [1] 1.0148481 -0.2446740 1.8991065 1.2629984 0.2379259 -0.3441057
tst<- Normlog2vsumasM3
Segundo histograma
hist(tst, breaks = nbreaks, col= 1:5,
main = 'Normalized Log2vsumasM3',
xlab='pEhEx1',
ylab= 'Frequency pEhEx')
Ajustando modelo
fw1<-fitdist(tst, "norm")
plotdist(tst, histo = TRUE, demp = TRUE)
nnorm.f <- fitdist(tst,"norm")
summary(nnorm.f)
## Fitting of the distribution ' norm ' by maximum likelihood
## Parameters :
## estimate Std. Error
## mean -2.967667e-17 0.01425665
## sd 9.998983e-01 0.01008093
## Loglikelihood: -6979.259 AIC: 13962.52 BIC: 13975.52
## Correlation matrix:
## mean sd
## mean 1 0
## sd 0 1
par(mfrow=c(2,2))
denscomp(nnorm.f,legendtext = 'Dist Normal')
qqcomp(nnorm.f,legendtext = 'Dist Normal')
cdfcomp(nnorm.f,legendtext = 'Dist Normal')
ppcomp(nnorm.f,legendtext = 'Dist Normal')
Calculo de cuantiles
probs <- c();
probs[8] = 0.175; probs[9] = 0.825;
probs[7] = 0.15; probs[10] = 0.85;
probs[6] = 0.125; probs[11] = 0.875;
probs[5] = 0.1; probs[12] = 0.9;
probs[4] = 0.075; probs[13] = 0.925;
probs[3] = 0.05; probs[14] = 0.95;
probs[2] = 0.025; probs[15] = 0.975;
probs[1] = 0.005; probs[16] = 0.995;
CuantilesData <- quantile(tst,prob = probs)
CuantilesModel <- qnorm(probs, mean=0, sd=1)
Cuantilillos <- t(CuantilesModel)
colnames(Cuantilillos) <- c('0.5%','2.5%','5%','7.5%',
'10%','12.5%','15%','17.5%',
'82.5%','85%','87.5%','90%',
'92.5%','95%','97.5%','99.5%')
Cuantilillos <- t(Cuantilillos)
colnames(Cuantilillos) <- c('Cuantiles Ajuste')
print(Cuantilillos)
## Cuantiles Ajuste
## 0.5% -2.5758293
## 2.5% -1.9599640
## 5% -1.6448536
## 7.5% -1.4395315
## 10% -1.2815516
## 12.5% -1.1503494
## 15% -1.0364334
## 17.5% -0.9345893
## 82.5% 0.9345893
## 85% 1.0364334
## 87.5% 1.1503494
## 90% 1.2815516
## 92.5% 1.4395315
## 95% 1.6448536
## 97.5% 1.9599640
## 99.5% 2.5758293
CuantilesA <- matrix(0,8,2)
CuantilesD <- matrix(0,8,2)
colnames(CuantilesA) <- c('LimInf','LimSup')
colnames(CuantilesD) <- c('LimInf','LimSup')
rownames(CuantilesA) <- c('65','70','75','80','85','90','95','99')
rownames(CuantilesD) <- c('65','70','75','80','85','90','95','99')
CuantilesA[1,1] <-CuantilesData[8]; CuantilesA[1,2] <-CuantilesData[9]
CuantilesA[2,1] <-CuantilesData[7]; CuantilesA[2,2] <-CuantilesData[10]
CuantilesA[3,1] <-CuantilesData[6]; CuantilesA[3,2] <-CuantilesData[11]
CuantilesA[4,1] <-CuantilesData[5]; CuantilesA[4,2] <-CuantilesData[12]
CuantilesA[5,1] <-CuantilesData[4]; CuantilesA[5,2] <-CuantilesData[13]
CuantilesA[6,1] <-CuantilesData[3]; CuantilesA[6,2] <-CuantilesData[14]
CuantilesA[7,1] <-CuantilesData[2]; CuantilesA[7,2] <-CuantilesData[15]
CuantilesA[8,1] <-CuantilesData[1]; CuantilesA[8,2] <-CuantilesData[16]
CuantilesD[1,1] <-Cuantilillos[8]; CuantilesD[1,2] <-Cuantilillos[9]
CuantilesD[2,1] <-Cuantilillos[7]; CuantilesD[2,2] <-Cuantilillos[10]
CuantilesD[3,1] <-Cuantilillos[6]; CuantilesD[3,2] <-Cuantilillos[11]
CuantilesD[4,1] <-Cuantilillos[5]; CuantilesD[4,2] <-Cuantilillos[12]
CuantilesD[5,1] <-Cuantilillos[4]; CuantilesD[5,2] <-Cuantilillos[13]
CuantilesD[6,1] <-Cuantilillos[3]; CuantilesD[6,2] <-Cuantilillos[14]
CuantilesD[7,1] <-Cuantilillos[2]; CuantilesD[7,2] <-Cuantilillos[15]
CuantilesD[8,1] <-Cuantilillos[1]; CuantilesD[8,2] <-Cuantilillos[16]
print(CuantilesD)
## LimInf LimSup
## 65 -0.9345893 0.9345893
## 70 -1.0364334 1.0364334
## 75 -1.1503494 1.1503494
## 80 -1.2815516 1.2815516
## 85 -1.4395315 1.4395315
## 90 -1.6448536 1.6448536
## 95 -1.9599640 1.9599640
## 99 -2.5758293 2.5758293
print(CuantilesA)
## LimInf LimSup
## 65 -0.8628952 0.8884901
## 70 -0.9319289 1.0207444
## 75 -1.0144701 1.1904247
## 80 -1.1171230 1.3814634
## 85 -1.2529569 1.6102025
## 90 -1.4542681 1.9460444
## 95 -1.8520407 2.4110529
## 99 -1.8520407 3.0342081
Histogramas
col_sequence <- rainbow(n = 7, alpha = 0.35, start = 0, end = 1)
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM3 pEhEx - DATA', lty = 9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesA[2,2], lty=2, col="darkblue") # 70% SUPERIOR
abline(v=CuantilesA[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesA[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesA[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesA[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesA[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesA[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesA[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesA[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesA[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesA[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM3 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[2,1], lty=2, col="darkblue"); # 70% INFERIOR
abline(v=CuantilesD[2,2], lty=2, col="darkblue"); # 70% SUPERIOR
abline(v=CuantilesD[3,1], lty=2, col="aquamarine4"); # 75% INFERIOR
abline(v=CuantilesD[3,2], lty=2, col="aquamarine4"); # 75% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
abline(v=CuantilesD[5,1], lty=2, col="brown"); # 85% INFERIOR
abline(v=CuantilesD[5,2], lty=2, col="brown"); # 85% SUPERIOR
abline(v=CuantilesD[6,1], lty=2, col="red"); # 90% INFERIOR
abline(v=CuantilesD[6,2], lty=2, col="red"); # 90% SUPERIOR
abline(v=CuantilesD[7,1], lty=2, col="blue"); # 95% INFERIOR
abline(v=CuantilesD[7,2], lty=2, col="blue"); # 95% SUPERIOR
abline(v=CuantilesD[8,1], lty=2, col="orange"); # 99% INFERIOR
abline(v=CuantilesD[8,2], lty=2, col="orange"); # 99% SUPERIOR
legend("topright",
legend=c("65%","70%","75%","80%","85%","90%","95%","99%"),
pch=c(1,2,3,4,5,6,7,8),
col=c("darkgoldenrod4","darkblue","aquamarine4",
"green", "brown","red","blue","orange"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM3 pEhEx - DATA', lty=9)
abline(v=CuantilesA[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesA[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesA[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesA[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM1 - ADJUSTED', lty=9)
abline(v=CuantilesD[1,1], lty=2, col="darkgoldenrod4"); # 65% INFERIOR
abline(v=CuantilesD[1,2], lty=2, col="darkgoldenrod4"); # 65% SUPERIOR
abline(v=CuantilesD[4,1], lty=2, col="green"); # 80% INFERIOR
abline(v=CuantilesD[4,2], lty=2, col="green"); # 80% SUPERIOR
legend("topright",legend=c("65%","80%"),
pch=c(1,2),col=c("darkgoldenrod4","green"))
par(mfrow=c(2,1))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM3 pEhEx - DATA', lty=9)
abline(v=CuantilesA[2,1], lty=2, col="darkblue");
abline(v=CuantilesA[2,2], lty=2, col="darkblue")
abline(v=CuantilesA[5,1], lty=2, col="brown");
abline(v=CuantilesA[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))
hist(tst, breaks = nbreaks, col = col_sequence,
main = 'Normalized Log2vsumasM3 pEhEx - ADJUSTED', lty=9)
abline(v=CuantilesD[2,1], lty=2, col="darkblue");
abline(v=CuantilesD[2,2], lty=2, col="darkblue")
abline(v=CuantilesD[5,1], lty=2, col="brown");
abline(v=CuantilesD[5,2], lty=2, col="brown")
legend("topright",legend=c("70%","85%"),
pch=c(1,2),#3,4,5,6,7,8),
col=c("brown"))